The current research investigates the effectiveness of a 16- session family therapy with adolescent substance abusers and their families. This competing continuation proposal seeks funds to complete this project, which has been operational since July 1, 1985. Funding was recommended for the first two of the proposed five years in order that feasibility in the areas of recruitment, random assignment and treatment might be demonstrated. We have been successful in these areas. The project is recruiting, randomly assigning, and treating subjects at several San Francisco Bay Area sites. Specifically, this study compares Structural-Strategic family therapy (SSFT) with a Multi-Family Education approach, and Group Therapy. Subjects are randomly assigned to one of these conditions, and after attrition, data will be collected on a total of 150 families (50 in each condition). Each treatment has a total time of 16 hours and all project providers have comparable levels of experience. Each category of treatment provider received structured training, and ongoing supervision, in their treatment condition. Instrumentation includes biochemical (urinalysis), self-report, perception of other, and family interaction tasks to measure change in individual, subsystem and family functioning. There are five assessment points: pre, during (for the adolescent's urinalysis), post treatment, and at 6 months, one, and two years after treatment. All three experimental conditions aim to ameliorate the symptom of drug abuse and prevent drug use in the younger sibling. A key assumption of SSFT is that a change in family functioning yields a positive change with the adolescent's substance abuse and reduces the younger siblings' substance abuse risk. The study includes a comparison sample of 50 non-clinical families with adolescents to allow a precise comparison of individual and interactional differences between families who have an adolescent substance abuser and those who do not. A multiple regression data analytic strategy will be employed: slope analyses will be computed in order to test the effects of the three experimental conditions. MANOVA will be used to test whether and how families of adolescent drug abusers differ from the non- clinical, comparison group.